## [1] "2024-04-11 11:37:52 CEST"
## [1] "explicated variable of regression : rh98"
## [1] "for Guinean_forest-savanna_regression_rh98.RDS"
## [2] "for Northern_Congolian_Forest-Savanna_regression_rh98.RDS"
## [3] "for Sahelian_Acacia_savanna_regression_rh98.RDS"
## [4] "for Southern_Congolian_forest-savanna_regression_rh98.RDS"
## [5] "for West_Sudanian_savanna_regression_rh98.RDS"
## [6] "for Western_Congolian_forest-savanna_regression_rh98.RDS"
## [1] "########################################"
## [1] "########################################"
## [1] "########################################"
## [1] "below, stancode for Guinean_forest-savanna_regression_rh98.RDS"
## // generated with brms 2.20.4
## functions {
## }
## data {
## int<lower=1> N; // total number of observations
## vector[N] Y; // response variable
## int<lower=1> K; // number of population-level effects
## matrix[N, K] X; // population-level design matrix
## int<lower=1> Kc; // number of population-level effects after centering
## int prior_only; // should the likelihood be ignored?
## }
## transformed data {
## matrix[N, Kc] Xc; // centered version of X without an intercept
## vector[Kc] means_X; // column means of X before centering
## for (i in 2:K) {
## means_X[i - 1] = mean(X[, i]);
## Xc[, i - 1] = X[, i] - means_X[i - 1];
## }
## }
## parameters {
## vector[Kc] b; // regression coefficients
## real Intercept; // temporary intercept for centered predictors
## real<lower=0> shape; // shape parameter
## }
## transformed parameters {
## real lprior = 0; // prior contributions to the log posterior
## lprior += student_t_lpdf(Intercept | 3, 2, 2.5);
## lprior += gamma_lpdf(shape | 0.01, 0.01);
## }
## model {
## // likelihood including constants
## if (!prior_only) {
## // initialize linear predictor term
## vector[N] mu = rep_vector(0.0, N);
## mu += Intercept + Xc * b;
## mu = exp(mu);
## target += gamma_lpdf(Y | shape, shape ./ mu);
## }
## // priors including constants
## target += lprior;
## }
## generated quantities {
## // actual population-level intercept
## real b_Intercept = Intercept - dot_product(means_X, b);
## }
## [1] "########################################"
## [1] "########################################"
## [1] "########################################"
## [1] " "
## [1] " "
## [1] "Guinean_forest-savanna_regression_rh98.RDS"
## [1] "########################################"
## Family: gamma
## Links: mu = log; shape = identity
## Formula: rh98 ~ fire_freq_std + mean_precip_std
## Data: table_region (Number of observations: 1725)
## Draws: 4 chains, each with iter = 10000; warmup = 2000; thin = 10;
## total post-warmup draws = 3200
##
## Population-Level Effects:
## Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
## Intercept 1.96 0.03 1.91 2.02 1.00 3075 3028
## fire_freq_std 0.03 0.01 0.01 0.05 1.00 3072 3045
## mean_precip_std 0.06 0.01 0.04 0.09 1.00 3046 3023
##
## Family Specific Parameters:
## Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
## shape 3.80 0.12 3.56 4.04 1.00 3076 3171
##
## Draws were sampled using sampling(NUTS). For each parameter, Bulk_ESS
## and Tail_ESS are effective sample size measures, and Rhat is the potential
## scale reduction factor on split chains (at convergence, Rhat = 1).
## [1] "########################################"
## [1] "Northern_Congolian_Forest-Savanna_regression_rh98.RDS"
## [1] "########################################"
## Family: gamma
## Links: mu = log; shape = identity
## Formula: rh98 ~ fire_freq_std + mean_precip_std
## Data: table_region (Number of observations: 243)
## Draws: 4 chains, each with iter = 10000; warmup = 2000; thin = 10;
## total post-warmup draws = 3200
##
## Population-Level Effects:
## Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
## Intercept 2.13 0.19 1.76 2.49 1.00 3075 3089
## fire_freq_std -0.03 0.03 -0.09 0.04 1.00 3042 3003
## mean_precip_std -0.00 0.09 -0.17 0.17 1.00 3065 3131
##
## Family Specific Parameters:
## Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
## shape 5.01 0.45 4.15 5.91 1.00 3021 3166
##
## Draws were sampled using sampling(NUTS). For each parameter, Bulk_ESS
## and Tail_ESS are effective sample size measures, and Rhat is the potential
## scale reduction factor on split chains (at convergence, Rhat = 1).
## [1] "########################################"
## [1] "Sahelian_Acacia_savanna_regression_rh98.RDS"
## [1] "########################################"
## Family: gamma
## Links: mu = log; shape = identity
## Formula: rh98 ~ fire_freq_std + mean_precip_std
## Data: table_region (Number of observations: 5563)
## Draws: 4 chains, each with iter = 10000; warmup = 2000; thin = 10;
## total post-warmup draws = 3200
##
## Population-Level Effects:
## Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
## Intercept 1.32 0.01 1.31 1.33 1.00 3297 3124
## fire_freq_std 0.11 0.01 0.10 0.12 1.00 3220 3216
## mean_precip_std 0.33 0.01 0.31 0.35 1.00 3299 3134
##
## Family Specific Parameters:
## Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
## shape 18.01 0.34 17.34 18.69 1.00 3452 3136
##
## Draws were sampled using sampling(NUTS). For each parameter, Bulk_ESS
## and Tail_ESS are effective sample size measures, and Rhat is the potential
## scale reduction factor on split chains (at convergence, Rhat = 1).
## [1] "########################################"
## [1] "Southern_Congolian_forest-savanna_regression_rh98.RDS"
## [1] "########################################"
## Family: gamma
## Links: mu = log; shape = identity
## Formula: rh98 ~ fire_freq_std + mean_precip_std
## Data: table_region (Number of observations: 47)
## Draws: 4 chains, each with iter = 10000; warmup = 2000; thin = 10;
## total post-warmup draws = 3200
##
## Population-Level Effects:
## Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
## Intercept 3.37 0.65 2.10 4.68 1.00 3137 3092
## fire_freq_std -0.08 0.04 -0.16 0.01 1.00 2968 3183
## mean_precip_std -0.96 0.31 -1.57 -0.36 1.00 3195 3092
##
## Family Specific Parameters:
## Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
## shape 6.07 1.25 3.88 8.75 1.00 3131 2904
##
## Draws were sampled using sampling(NUTS). For each parameter, Bulk_ESS
## and Tail_ESS are effective sample size measures, and Rhat is the potential
## scale reduction factor on split chains (at convergence, Rhat = 1).
## [1] "########################################"
## [1] "West_Sudanian_savanna_regression_rh98.RDS"
## [1] "########################################"
## Family: gamma
## Links: mu = log; shape = identity
## Formula: rh98 ~ fire_freq_std + mean_precip_std
## Data: table_region (Number of observations: 3277)
## Draws: 4 chains, each with iter = 10000; warmup = 2000; thin = 10;
## total post-warmup draws = 3200
##
## Population-Level Effects:
## Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
## Intercept 1.22 0.02 1.19 1.25 1.00 3067 3114
## fire_freq_std 0.08 0.01 0.07 0.10 1.00 3044 3056
## mean_precip_std 0.53 0.02 0.50 0.56 1.00 3052 2874
##
## Family Specific Parameters:
## Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
## shape 5.47 0.13 5.21 5.73 1.00 3600 2875
##
## Draws were sampled using sampling(NUTS). For each parameter, Bulk_ESS
## and Tail_ESS are effective sample size measures, and Rhat is the potential
## scale reduction factor on split chains (at convergence, Rhat = 1).
## [1] "########################################"
## [1] "Western_Congolian_forest-savanna_regression_rh98.RDS"
## [1] "########################################"
## Family: gamma
## Links: mu = log; shape = identity
## Formula: rh98 ~ fire_freq_std + mean_precip_std
## Data: table_region (Number of observations: 259)
## Draws: 4 chains, each with iter = 10000; warmup = 2000; thin = 10;
## total post-warmup draws = 3200
##
## Population-Level Effects:
## Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
## Intercept 1.36 0.16 1.05 1.66 1.00 2982 3193
## fire_freq_std -0.05 0.02 -0.10 -0.00 1.00 2648 2861
## mean_precip_std 0.36 0.10 0.18 0.56 1.00 3026 3172
##
## Family Specific Parameters:
## Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
## shape 3.07 0.26 2.61 3.60 1.00 3508 3132
##
## Draws were sampled using sampling(NUTS). For each parameter, Bulk_ESS
## and Tail_ESS are effective sample size measures, and Rhat is the potential
## scale reduction factor on split chains (at convergence, Rhat = 1).
## [1] "########################################"
## [1] "explicated variable of regression : canopy_cover"
## [1] "for Guinean_forest-savanna_regression_canopy_cover.RDS"
## [2] "for Northern_Congolian_Forest-Savanna_regression_canopy_cover.RDS"
## [3] "for Sahelian_Acacia_savanna_regression_canopy_cover.RDS"
## [4] "for Southern_Congolian_forest-savanna_regression_canopy_cover.RDS"
## [5] "for West_Sudanian_savanna_regression_canopy_cover.RDS"
## [6] "for Western_Congolian_forest-savanna_regression_canopy_cover.RDS"
## [1] "########################################"
## [1] "########################################"
## [1] "########################################"
## [1] "below, stancode for Guinean_forest-savanna_regression_canopy_cover.RDS"
## // generated with brms 2.20.4
## functions {
## /* zero-inflated beta log-PDF of a single response
## * Args:
## * y: the response value
## * mu: mean parameter of the beta distribution
## * phi: precision parameter of the beta distribution
## * zi: zero-inflation probability
## * Returns:
## * a scalar to be added to the log posterior
## */
## real zero_inflated_beta_lpdf(real y, real mu, real phi, real zi) {
## row_vector[2] shape = [mu * phi, (1 - mu) * phi];
## if (y == 0) {
## return bernoulli_lpmf(1 | zi);
## } else {
## return bernoulli_lpmf(0 | zi) +
## beta_lpdf(y | shape[1], shape[2]);
## }
## }
## /* zero-inflated beta log-PDF of a single response
## * logit parameterization of the zero-inflation part
## * Args:
## * y: the response value
## * mu: mean parameter of the beta distribution
## * phi: precision parameter of the beta distribution
## * zi: linear predictor for zero-inflation part
## * Returns:
## * a scalar to be added to the log posterior
## */
## real zero_inflated_beta_logit_lpdf(real y, real mu, real phi, real zi) {
## row_vector[2] shape = [mu * phi, (1 - mu) * phi];
## if (y == 0) {
## return bernoulli_logit_lpmf(1 | zi);
## } else {
## return bernoulli_logit_lpmf(0 | zi) +
## beta_lpdf(y | shape[1], shape[2]);
## }
## }
## // zero-inflated beta log-CCDF and log-CDF functions
## real zero_inflated_beta_lccdf(real y, real mu, real phi, real zi) {
## row_vector[2] shape = [mu * phi, (1 - mu) * phi];
## return bernoulli_lpmf(0 | zi) + beta_lccdf(y | shape[1], shape[2]);
## }
## real zero_inflated_beta_lcdf(real y, real mu, real phi, real zi) {
## return log1m_exp(zero_inflated_beta_lccdf(y | mu, phi, zi));
## }
## }
## data {
## int<lower=1> N; // total number of observations
## vector[N] Y; // response variable
## int<lower=1> K; // number of population-level effects
## matrix[N, K] X; // population-level design matrix
## int<lower=1> Kc; // number of population-level effects after centering
## int prior_only; // should the likelihood be ignored?
## }
## transformed data {
## matrix[N, Kc] Xc; // centered version of X without an intercept
## vector[Kc] means_X; // column means of X before centering
## for (i in 2:K) {
## means_X[i - 1] = mean(X[, i]);
## Xc[, i - 1] = X[, i] - means_X[i - 1];
## }
## }
## parameters {
## vector[Kc] b; // regression coefficients
## real Intercept; // temporary intercept for centered predictors
## real<lower=0> phi; // precision parameter
## real<lower=0,upper=1> zi; // zero-inflation probability
## }
## transformed parameters {
## real lprior = 0; // prior contributions to the log posterior
## lprior += student_t_lpdf(Intercept | 3, 0, 2.5);
## lprior += gamma_lpdf(phi | 0.01, 0.01);
## lprior += beta_lpdf(zi | 1, 1);
## }
## model {
## // likelihood including constants
## if (!prior_only) {
## // initialize linear predictor term
## vector[N] mu = rep_vector(0.0, N);
## mu += Intercept + Xc * b;
## mu = inv_logit(mu);
## for (n in 1:N) {
## target += zero_inflated_beta_lpdf(Y[n] | mu[n], phi, zi);
## }
## }
## // priors including constants
## target += lprior;
## }
## generated quantities {
## // actual population-level intercept
## real b_Intercept = Intercept - dot_product(means_X, b);
## }
## [1] "########################################"
## [1] "########################################"
## [1] "########################################"
## [1] " "
## [1] " "
## [1] "Guinean_forest-savanna_regression_canopy_cover.RDS"
## [1] "########################################"
## Family: zero_inflated_beta
## Links: mu = logit; phi = identity; zi = identity
## Formula: canopy_cover ~ fire_freq_std + mean_precip_std
## Data: table_region (Number of observations: 1725)
## Draws: 3 chains, each with iter = 10000; warmup = 2000; thin = 10;
## total post-warmup draws = 2400
##
## Population-Level Effects:
## Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
## Intercept -1.95 0.06 -2.06 -1.84 1.00 2329 2393
## fire_freq_std -0.03 0.02 -0.07 0.02 1.00 2512 2410
## mean_precip_std 0.12 0.03 0.07 0.17 1.00 2246 2185
##
## Family Specific Parameters:
## Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
## phi 5.28 0.19 4.93 5.68 1.00 2300 2519
## zi 0.11 0.01 0.10 0.13 1.00 2349 2292
##
## Draws were sampled using sampling(NUTS). For each parameter, Bulk_ESS
## and Tail_ESS are effective sample size measures, and Rhat is the potential
## scale reduction factor on split chains (at convergence, Rhat = 1).
## [1] "########################################"
## [1] "Northern_Congolian_Forest-Savanna_regression_canopy_cover.RDS"
## [1] "########################################"
## Family: zero_inflated_beta
## Links: mu = logit; phi = identity; zi = identity
## Formula: canopy_cover ~ fire_freq_std + mean_precip_std
## Data: table_region (Number of observations: 243)
## Draws: 3 chains, each with iter = 10000; warmup = 2000; thin = 10;
## total post-warmup draws = 2400
##
## Population-Level Effects:
## Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
## Intercept -1.77 0.36 -2.49 -1.06 1.00 2062 2093
## fire_freq_std -0.00 0.06 -0.13 0.11 1.00 2320 2300
## mean_precip_std 0.10 0.17 -0.24 0.41 1.00 2021 2090
##
## Family Specific Parameters:
## Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
## phi 6.72 0.62 5.57 7.98 1.00 2172 2174
## zi 0.06 0.02 0.03 0.09 1.00 2462 2175
##
## Draws were sampled using sampling(NUTS). For each parameter, Bulk_ESS
## and Tail_ESS are effective sample size measures, and Rhat is the potential
## scale reduction factor on split chains (at convergence, Rhat = 1).
## [1] "########################################"
## [1] "Sahelian_Acacia_savanna_regression_canopy_cover.RDS"
## [1] "########################################"
## Family: zero_inflated_beta
## Links: mu = logit; phi = identity; zi = identity
## Formula: canopy_cover ~ fire_freq_std + mean_precip_std
## Data: table_region (Number of observations: 5563)
## Draws: 3 chains, each with iter = 10000; warmup = 2000; thin = 10;
## total post-warmup draws = 2400
##
## Population-Level Effects:
## Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
## Intercept -3.89 0.04 -3.96 -3.81 1.00 2460 2389
## fire_freq_std 0.18 0.03 0.12 0.23 1.00 2266 2215
## mean_precip_std 0.85 0.06 0.73 0.97 1.00 2353 2288
##
## Family Specific Parameters:
## Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
## phi 27.54 1.24 25.14 29.96 1.00 2436 2332
## zi 0.68 0.01 0.67 0.69 1.00 2484 2090
##
## Draws were sampled using sampling(NUTS). For each parameter, Bulk_ESS
## and Tail_ESS are effective sample size measures, and Rhat is the potential
## scale reduction factor on split chains (at convergence, Rhat = 1).
## [1] "########################################"
## [1] "Southern_Congolian_forest-savanna_regression_canopy_cover.RDS"
## [1] "########################################"
## Family: zero_inflated_beta
## Links: mu = logit; phi = identity; zi = identity
## Formula: canopy_cover ~ fire_freq_std + mean_precip_std
## Data: table_region (Number of observations: 47)
## Draws: 3 chains, each with iter = 10000; warmup = 2000; thin = 10;
## total post-warmup draws = 2400
##
## Population-Level Effects:
## Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
## Intercept -1.68 2.43 -6.50 3.17 1.00 2565 2352
## fire_freq_std -0.04 0.15 -0.34 0.25 1.00 2875 2352
## mean_precip_std -0.20 1.18 -2.55 2.05 1.00 2539 2409
##
## Family Specific Parameters:
## Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
## phi 4.47 1.51 2.01 7.78 1.00 2263 2328
## zi 0.51 0.07 0.38 0.64 1.00 2388 2116
##
## Draws were sampled using sampling(NUTS). For each parameter, Bulk_ESS
## and Tail_ESS are effective sample size measures, and Rhat is the potential
## scale reduction factor on split chains (at convergence, Rhat = 1).
## [1] "########################################"
## [1] "West_Sudanian_savanna_regression_canopy_cover.RDS"
## [1] "########################################"
## Family: zero_inflated_beta
## Links: mu = logit; phi = identity; zi = identity
## Formula: canopy_cover ~ fire_freq_std + mean_precip_std
## Data: table_region (Number of observations: 3277)
## Draws: 3 chains, each with iter = 10000; warmup = 2000; thin = 10;
## total post-warmup draws = 2400
##
## Population-Level Effects:
## Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
## Intercept -3.52 0.05 -3.61 -3.42 1.00 2378 2285
## fire_freq_std 0.08 0.01 0.05 0.11 1.00 2402 2369
## mean_precip_std 0.90 0.04 0.82 0.98 1.00 2507 2256
##
## Family Specific Parameters:
## Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
## phi 9.97 0.32 9.36 10.63 1.00 2315 2251
## zi 0.23 0.01 0.21 0.24 1.00 2245 2288
##
## Draws were sampled using sampling(NUTS). For each parameter, Bulk_ESS
## and Tail_ESS are effective sample size measures, and Rhat is the potential
## scale reduction factor on split chains (at convergence, Rhat = 1).
## [1] "########################################"
## [1] "Western_Congolian_forest-savanna_regression_canopy_cover.RDS"
## [1] "########################################"
## Family: zero_inflated_beta
## Links: mu = logit; phi = identity; zi = identity
## Formula: canopy_cover ~ fire_freq_std + mean_precip_std
## Data: table_region (Number of observations: 259)
## Draws: 3 chains, each with iter = 10000; warmup = 2000; thin = 10;
## total post-warmup draws = 2400
##
## Population-Level Effects:
## Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
## Intercept -1.91 0.33 -2.54 -1.25 1.00 2474 2412
## fire_freq_std -0.13 0.05 -0.23 -0.03 1.00 2465 2251
## mean_precip_std 0.20 0.21 -0.22 0.58 1.00 2423 2348
##
## Family Specific Parameters:
## Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
## phi 3.51 0.39 2.79 4.30 1.00 2551 2455
## zi 0.22 0.03 0.17 0.28 1.00 2147 2145
##
## Draws were sampled using sampling(NUTS). For each parameter, Bulk_ESS
## and Tail_ESS are effective sample size measures, and Rhat is the potential
## scale reduction factor on split chains (at convergence, Rhat = 1).
## [1] "########################################"
## [1] "Gamma regressions for Guinean_forest-savanna_regression_rh98.RDS"
## [2] "Gamma regressions for Northern_Congolian_Forest-Savanna_regression_rh98.RDS"
## [3] "Gamma regressions for Sahelian_Acacia_savanna_regression_rh98.RDS"
## [4] "Gamma regressions for Southern_Congolian_forest-savanna_regression_rh98.RDS"
## [5] "Gamma regressions for West_Sudanian_savanna_regression_rh98.RDS"
## [6] "Gamma regressions for Western_Congolian_forest-savanna_regression_rh98.RDS"
## [1] "########################################"
## [1] "########################################"
## [1] "########################################"
## [1] "Guinean_forest-savanna_regression_rh98.RDS"
## [1] "########################################"
## [1] "dim(linear_predictors_for_one_beta_draw_per_column)"
## [1] 1725 3200
## [1] "(nb_donnes I * nb_iter_mcmc J)"
## [1] "mean(table_region$rh98)"
## [1] 8.17064
## [1] "sd(table_region$rh98)"
## [1] 4.228092
## [1] "mean(simulations[,j]) ( truth = 8.171 )"
## [1] 8.253
## [1] "sd(simulations[,j]) ( truth = 4.228 )"
## [1] 4.327
## [1] "mean(simulations[,j]) ( truth = 8.171 )"
## [1] 8.154
## [1] "sd(simulations[,j]) ( truth = 4.228 )"
## [1] 4.026
## [1] "mean(simulations[,j]) ( truth = 8.171 )"
## [1] 8.224
## [1] "sd(simulations[,j]) ( truth = 4.228 )"
## [1] 4.332
## [1] "mean(simulations[,j]) ( truth = 8.171 )"
## [1] 8.346
## [1] "sd(simulations[,j]) ( truth = 4.228 )"
## [1] 4.438
## [1] "mean(simulations[,j]) ( truth = 8.171 )"
## [1] 8.113
## [1] "sd(simulations[,j]) ( truth = 4.228 )"
## [1] 4.289
## [1] "Northern_Congolian_Forest-Savanna_regression_rh98.RDS"
## [1] "########################################"
## [1] "dim(linear_predictors_for_one_beta_draw_per_column)"
## [1] 243 3200
## [1] "(nb_donnes I * nb_iter_mcmc J)"
## [1] "mean(table_region$rh98)"
## [1] 8.251706
## [1] "sd(table_region$rh98)"
## [1] 3.52155
## [1] "mean(simulations[,j]) ( truth = 8.252 )"
## [1] 8.047
## [1] "sd(simulations[,j]) ( truth = 3.522 )"
## [1] 3.888
## [1] "mean(simulations[,j]) ( truth = 8.252 )"
## [1] 8.381
## [1] "sd(simulations[,j]) ( truth = 3.522 )"
## [1] 3.469
## [1] "mean(simulations[,j]) ( truth = 8.252 )"
## [1] 8.355
## [1] "sd(simulations[,j]) ( truth = 3.522 )"
## [1] 3.233
## [1] "mean(simulations[,j]) ( truth = 8.252 )"
## [1] 8.365
## [1] "sd(simulations[,j]) ( truth = 3.522 )"
## [1] 3.437
## [1] "mean(simulations[,j]) ( truth = 8.252 )"
## [1] 7.952
## [1] "sd(simulations[,j]) ( truth = 3.522 )"
## [1] 3.517
## [1] "Sahelian_Acacia_savanna_regression_rh98.RDS"
## [1] "########################################"
## [1] "dim(linear_predictors_for_one_beta_draw_per_column)"
## [1] 5563 3200
## [1] "(nb_donnes I * nb_iter_mcmc J)"
## [1] "mean(table_region$rh98)"
## [1] 3.049537
## [1] "sd(table_region$rh98)"
## [1] 1.187586
## [1] "mean(simulations[,j]) ( truth = 3.05 )"
## [1] 3.033
## [1] "sd(simulations[,j]) ( truth = 1.188 )"
## [1] 0.931
## [1] "mean(simulations[,j]) ( truth = 3.05 )"
## [1] 3.032
## [1] "sd(simulations[,j]) ( truth = 1.188 )"
## [1] 0.95
## [1] "mean(simulations[,j]) ( truth = 3.05 )"
## [1] 3.045
## [1] "sd(simulations[,j]) ( truth = 1.188 )"
## [1] 1
## [1] "mean(simulations[,j]) ( truth = 3.05 )"
## [1] 3.051
## [1] "sd(simulations[,j]) ( truth = 1.188 )"
## [1] 0.976
## [1] "mean(simulations[,j]) ( truth = 3.05 )"
## [1] 3.055
## [1] "sd(simulations[,j]) ( truth = 1.188 )"
## [1] 0.976
## [1] "Southern_Congolian_forest-savanna_regression_rh98.RDS"
## [1] "########################################"
## [1] "dim(linear_predictors_for_one_beta_draw_per_column)"
## [1] 47 3200
## [1] "(nb_donnes I * nb_iter_mcmc J)"
## [1] "mean(table_region$rh98)"
## [1] 3.938866
## [1] "sd(table_region$rh98)"
## [1] 2.086246
## [1] "mean(simulations[,j]) ( truth = 3.939 )"
## [1] 4.288
## [1] "sd(simulations[,j]) ( truth = 2.086 )"
## [1] 3.092
## [1] "mean(simulations[,j]) ( truth = 3.939 )"
## [1] 4.025
## [1] "sd(simulations[,j]) ( truth = 2.086 )"
## [1] 1.735
## [1] "mean(simulations[,j]) ( truth = 3.939 )"
## [1] 3.939
## [1] "sd(simulations[,j]) ( truth = 2.086 )"
## [1] 1.369
## [1] "mean(simulations[,j]) ( truth = 3.939 )"
## [1] 3.363
## [1] "sd(simulations[,j]) ( truth = 2.086 )"
## [1] 1.694
## [1] "mean(simulations[,j]) ( truth = 3.939 )"
## [1] 4.106
## [1] "sd(simulations[,j]) ( truth = 2.086 )"
## [1] 2.062
## [1] "West_Sudanian_savanna_regression_rh98.RDS"
## [1] "########################################"
## [1] "dim(linear_predictors_for_one_beta_draw_per_column)"
## [1] 3277 3200
## [1] "(nb_donnes I * nb_iter_mcmc J)"
## [1] "mean(table_region$rh98)"
## [1] 5.728088
## [1] "sd(table_region$rh98)"
## [1] 3.15554
## [1] "mean(simulations[,j]) ( truth = 5.728 )"
## [1] 5.827
## [1] "sd(simulations[,j]) ( truth = 3.156 )"
## [1] 3.211
## [1] "mean(simulations[,j]) ( truth = 5.728 )"
## [1] 5.723
## [1] "sd(simulations[,j]) ( truth = 3.156 )"
## [1] 3.169
## [1] "mean(simulations[,j]) ( truth = 5.728 )"
## [1] 5.771
## [1] "sd(simulations[,j]) ( truth = 3.156 )"
## [1] 3.171
## [1] "mean(simulations[,j]) ( truth = 5.728 )"
## [1] 5.722
## [1] "sd(simulations[,j]) ( truth = 3.156 )"
## [1] 3.019
## [1] "mean(simulations[,j]) ( truth = 5.728 )"
## [1] 5.717
## [1] "sd(simulations[,j]) ( truth = 3.156 )"
## [1] 3.088
## [1] "Western_Congolian_forest-savanna_regression_rh98.RDS"
## [1] "########################################"
## [1] "dim(linear_predictors_for_one_beta_draw_per_column)"
## [1] 259 3200
## [1] "(nb_donnes I * nb_iter_mcmc J)"
## [1] "mean(table_region$rh98)"
## [1] 6.137749
## [1] "sd(table_region$rh98)"
## [1] 4.607916
## [1] "mean(simulations[,j]) ( truth = 6.138 )"
## [1] 6.194
## [1] "sd(simulations[,j]) ( truth = 4.608 )"
## [1] 3.968
## [1] "mean(simulations[,j]) ( truth = 6.138 )"
## [1] 6.453
## [1] "sd(simulations[,j]) ( truth = 4.608 )"
## [1] 4.018
## [1] "mean(simulations[,j]) ( truth = 6.138 )"
## [1] 6.139
## [1] "sd(simulations[,j]) ( truth = 4.608 )"
## [1] 3.565
## [1] "mean(simulations[,j]) ( truth = 6.138 )"
## [1] 5.528
## [1] "sd(simulations[,j]) ( truth = 4.608 )"
## [1] 3.481
## [1] "mean(simulations[,j]) ( truth = 6.138 )"
## [1] 6.431
## [1] "sd(simulations[,j]) ( truth = 4.608 )"
## [1] 4.292
## [1] "Beta regressions for Guinean_forest-savanna_regression_canopy_cover.RDS"
## [2] "Beta regressions for Northern_Congolian_Forest-Savanna_regression_canopy_cover.RDS"
## [3] "Beta regressions for Sahelian_Acacia_savanna_regression_canopy_cover.RDS"
## [4] "Beta regressions for Southern_Congolian_forest-savanna_regression_canopy_cover.RDS"
## [5] "Beta regressions for West_Sudanian_savanna_regression_canopy_cover.RDS"
## [6] "Beta regressions for Western_Congolian_forest-savanna_regression_canopy_cover.RDS"
## [1] "########################################"
## [1] "########################################"
## [1] "########################################"
## [1] "Guinean_forest-savanna_regression_canopy_cover.RDS"
## [1] "########################################"
## [1] "dim(linear_predictors_for_one_beta_draw_per_column)"
## [1] 1725 2400
## [1] "(nb_donnes I * nb_iter_mcmc J)"
## Le chargement a nécessité le package : gtools
##
## Attachement du package : 'gtools'
## Les objets suivants sont masqués depuis 'package:brms':
##
## ddirichlet, rdirichlet
## [1] "mean(table_region$canopy_cover)"
## [1] 0.1314814
## [1] "sd(table_region$canopy_cover)"
## [1] 0.14466
## [1] "mean(simulations[,j]) ( truth = 0.131 )"
## [1] 0.136
## [1] "sd(simulations[,j]) ( truth = 0.145 )"
## [1] 0.149
## [1] "mean(simulations[,j]) ( truth = 0.131 )"
## [1] 0.128
## [1] "sd(simulations[,j]) ( truth = 0.145 )"
## [1] 0.143
## [1] "mean(simulations[,j]) ( truth = 0.131 )"
## [1] 0.125
## [1] "sd(simulations[,j]) ( truth = 0.145 )"
## [1] 0.139
## [1] "mean(simulations[,j]) ( truth = 0.131 )"
## [1] 0.136
## [1] "sd(simulations[,j]) ( truth = 0.145 )"
## [1] 0.142
## [1] "mean(simulations[,j]) ( truth = 0.131 )"
## [1] 0.136
## [1] "sd(simulations[,j]) ( truth = 0.145 )"
## [1] 0.141
## [1] "Northern_Congolian_Forest-Savanna_regression_canopy_cover.RDS"
## [1] "########################################"
## [1] "dim(linear_predictors_for_one_beta_draw_per_column)"
## [1] 243 2400
## [1] "(nb_donnes I * nb_iter_mcmc J)"
## [1] "mean(table_region$canopy_cover)"
## [1] 0.1649144
## [1] "sd(table_region$canopy_cover)"
## [1] 0.1334132
## [1] "mean(simulations[,j]) ( truth = 0.165 )"
## [1] 0.161
## [1] "sd(simulations[,j]) ( truth = 0.133 )"
## [1] 0.144
## [1] "mean(simulations[,j]) ( truth = 0.165 )"
## [1] 0.184
## [1] "sd(simulations[,j]) ( truth = 0.133 )"
## [1] 0.168
## [1] "mean(simulations[,j]) ( truth = 0.165 )"
## [1] 0.141
## [1] "sd(simulations[,j]) ( truth = 0.133 )"
## [1] 0.125
## [1] "mean(simulations[,j]) ( truth = 0.165 )"
## [1] 0.174
## [1] "sd(simulations[,j]) ( truth = 0.133 )"
## [1] 0.151
## [1] "mean(simulations[,j]) ( truth = 0.165 )"
## [1] 0.155
## [1] "sd(simulations[,j]) ( truth = 0.133 )"
## [1] 0.131
## [1] "Sahelian_Acacia_savanna_regression_canopy_cover.RDS"
## [1] "########################################"
## [1] "dim(linear_predictors_for_one_beta_draw_per_column)"
## [1] 5563 2400
## [1] "(nb_donnes I * nb_iter_mcmc J)"
## [1] "mean(table_region$canopy_cover)"
## [1] 0.005161202
## [1] "sd(table_region$canopy_cover)"
## [1] 0.02202149
## [1] "mean(simulations[,j]) ( truth = 0.005 )"
## [1] 0.004
## [1] "sd(simulations[,j]) ( truth = 0.022 )"
## [1] 0.013
## [1] "mean(simulations[,j]) ( truth = 0.005 )"
## [1] 0.004
## [1] "sd(simulations[,j]) ( truth = 0.022 )"
## [1] 0.013
## [1] "mean(simulations[,j]) ( truth = 0.005 )"
## [1] 0.004
## [1] "sd(simulations[,j]) ( truth = 0.022 )"
## [1] 0.013
## [1] "mean(simulations[,j]) ( truth = 0.005 )"
## [1] 0.004
## [1] "sd(simulations[,j]) ( truth = 0.022 )"
## [1] 0.013
## [1] "mean(simulations[,j]) ( truth = 0.005 )"
## [1] 0.004
## [1] "sd(simulations[,j]) ( truth = 0.022 )"
## [1] 0.014
## [1] "Southern_Congolian_forest-savanna_regression_canopy_cover.RDS"
## [1] "########################################"
## [1] "dim(linear_predictors_for_one_beta_draw_per_column)"
## [1] 47 2400
## [1] "(nb_donnes I * nb_iter_mcmc J)"
## [1] "mean(table_region$canopy_cover)"
## [1] 0.04984602
## [1] "sd(table_region$canopy_cover)"
## [1] 0.1063901
## [1] "mean(simulations[,j]) ( truth = 0.05 )"
## [1] 0.058
## [1] "sd(simulations[,j]) ( truth = 0.106 )"
## [1] 0.13
## [1] "mean(simulations[,j]) ( truth = 0.05 )"
## [1] 0.059
## [1] "sd(simulations[,j]) ( truth = 0.106 )"
## [1] 0.107
## [1] "mean(simulations[,j]) ( truth = 0.05 )"
## [1] 0.057
## [1] "sd(simulations[,j]) ( truth = 0.106 )"
## [1] 0.098
## [1] "mean(simulations[,j]) ( truth = 0.05 )"
## [1] 0.072
## [1] "sd(simulations[,j]) ( truth = 0.106 )"
## [1] 0.114
## [1] "mean(simulations[,j]) ( truth = 0.05 )"
## [1] 0.068
## [1] "sd(simulations[,j]) ( truth = 0.106 )"
## [1] 0.115
## [1] "West_Sudanian_savanna_regression_canopy_cover.RDS"
## [1] "########################################"
## [1] "dim(linear_predictors_for_one_beta_draw_per_column)"
## [1] 3277 2400
## [1] "(nb_donnes I * nb_iter_mcmc J)"
## [1] "mean(table_region$canopy_cover)"
## [1] 0.05300694
## [1] "sd(table_region$canopy_cover)"
## [1] 0.08641946
## [1] "mean(simulations[,j]) ( truth = 0.053 )"
## [1] 0.057
## [1] "sd(simulations[,j]) ( truth = 0.086 )"
## [1] 0.082
## [1] "mean(simulations[,j]) ( truth = 0.053 )"
## [1] 0.05
## [1] "sd(simulations[,j]) ( truth = 0.086 )"
## [1] 0.073
## [1] "mean(simulations[,j]) ( truth = 0.053 )"
## [1] 0.056
## [1] "sd(simulations[,j]) ( truth = 0.086 )"
## [1] 0.082
## [1] "mean(simulations[,j]) ( truth = 0.053 )"
## [1] 0.05
## [1] "sd(simulations[,j]) ( truth = 0.086 )"
## [1] 0.074
## [1] "mean(simulations[,j]) ( truth = 0.053 )"
## [1] 0.053
## [1] "sd(simulations[,j]) ( truth = 0.086 )"
## [1] 0.076
## [1] "Western_Congolian_forest-savanna_regression_canopy_cover.RDS"
## [1] "########################################"
## [1] "dim(linear_predictors_for_one_beta_draw_per_column)"
## [1] 259 2400
## [1] "(nb_donnes I * nb_iter_mcmc J)"
## [1] "mean(table_region$canopy_cover)"
## [1] 0.1079882
## [1] "sd(table_region$canopy_cover)"
## [1] 0.1649516
## [1] "mean(simulations[,j]) ( truth = 0.108 )"
## [1] 0.105
## [1] "sd(simulations[,j]) ( truth = 0.165 )"
## [1] 0.164
## [1] "mean(simulations[,j]) ( truth = 0.108 )"
## [1] 0.116
## [1] "sd(simulations[,j]) ( truth = 0.165 )"
## [1] 0.158
## [1] "mean(simulations[,j]) ( truth = 0.108 )"
## [1] 0.137
## [1] "sd(simulations[,j]) ( truth = 0.165 )"
## [1] 0.191
## [1] "mean(simulations[,j]) ( truth = 0.108 )"
## [1] 0.116
## [1] "sd(simulations[,j]) ( truth = 0.165 )"
## [1] 0.145
## [1] "mean(simulations[,j]) ( truth = 0.108 )"
## [1] 0.105
## [1] "sd(simulations[,j]) ( truth = 0.165 )"
## [1] 0.143
